In an attempt to find a solution similar to the FDM 3D printers which would allow cost-effective and reliable additive manufacturing of metal components, this paper proposes a three-axis WAAM system capable of reliably printing small, near-net-shape metal objects. The system consists of gas metal arc (GMA) process equipment, a three-axis CNC positioning system, the interpass temperature control and forced cooling of the base plate and the deposit. The main challenge addressed is the minimisation of shape distortions caused by excessive heat accumulation when printing small objects. The interpass temperature control uses an IR pyrometer to remotely measure the last deposited layer and a control system to keep the interpass temperature below the predefined value by stopping the deposition after each layer in order to allow the deposit to cool. This results in a stable and more repeatable shape of the deposit, even when the heat transfer conditions are changing during the build-up process. The combination of adaptive interlayer dwell time and forced cooling significantly improves system productivity. Open-source NC control and path generation software is used, which enables fast and easy creation of the control code. Different control methods are evaluated through the printing of simple walls, and the printing accuracy is evaluated by printing small shell objects. As the results show, the interpass temperature control allows small objects to be printed at near-net shape with a deviation of 2%, which means that successful printing of 3D shapes can be achieved without trial and error approach.

Efficient workspace awareness is critical for improved interaction in cooperative and collaborative robotic applications. In addition to safety and control aspects, quality-related tasks such as the monitoring of manual activities and the final quality assessment of the results are also required. In this context, a visual quality and safety monitoring system is developed and evaluated. The system integrates close-up observation of manual activities and posture monitoring. A compact single-camera stereo vision system and a time-of-flight depth camera are used to minimize the interference of the sensors with the operator and the workplace. Data processing is based on a deep learning to detect classes related to quality and safety aspects. The operation of the system is evaluated while monitoring a human-robot manual assembly task. The results show that the system ensures a high level of safety, provides reliable visual feedback to the operator on errors in the assembly process, and inspects the finished assembly with a low critical error rate.